Fluid responsiveness predicted by elevation of PEEP in patients with septic shock

The assessment of whether a patient is fluid responsive can be difficult in clinical practice. Invasive filling pressures are inadequate indicators of preload and fluid responsiveness in critically ill patients. Dynamic indices may be unreliable in clinical practice because of arrhythmias or spontaneous breathing efforts. Elevation of positive end-expiratory pressure (PEEP) causes cardiorespiratory interactions, which may produce signs of hypovolaemia. Our aim was to assess whether haemodynamic changes during a short elevation of PEEP would predict fluid responsiveness in patients with septic shock.

We performed a prospective observational study in 20 patients with septic shock on mechanical ventilation. We assessed the following changes in haemodynamic variables during a temporary elevation of PEEP from 10 cm H2O to 20 cm H2O during an end-expiratory pause: mean arterial pressure (MAP), systolic arterial pressure, pulse pressure, central venous pressure, pulmonary artery occlusion pressure, left ventricular end diastolic area and aortic velocity-time integral. We defined fluid responsiveness as an increase in cardiac output of 15% to a subsequent fluid challenge.

Decrease in MAP related to elevation of PEEP predicted fluid responsiveness (P = 0.003). The best cut-off value of ΔMAP for clinical use was -8%, with a negative predictive value for fluid responsiveness of 100%.

In patients with septic shock, the absence of decrease in MAP during an elevation of PEEP may be used to identify patients who will not increase their cardiac output in response to fluid challenge.